Background: The efficacy of dexamethasone in extending the duration of local anaesthetic block is uncertain. In a randomised controlled triple blind crossover study in volunteers, we tested the hypothesis that neither i.v. nor perineurally administered dexamethasone prolongs the sensory block achieved with ropivacaine. Methods: Ultrasound-guided ulnar nerve blocks (ropivacaine 0.75% wt/vol, 3 ml, with saline 1 ml with or without dexamethasone 4 mg) were performed on three occasions in 24 male volunteers along with an i.v. injection of saline 1 ml with or without dexamethasone 4 mg. The combinations of saline and dexamethasone were as follows: control group, perineural and i.v. saline; perineural group, perineural dexamethasone and i.v. saline; i.v. group, perineural saline and i.v. dexamethasone. Sensory block was measured using a VAS in response to pinprick testing. The duration of sensory block was the primary outcome and time to onset of sensory block the secondary outcome. Results: All 24 subjects completed the trial. The median [inter-quartile range (IQR)] duration of sensory block was 6.87 (5.85e7.62) h in the control group, 7.37 (5.78e7.93) h in the perineural group and 7.37 (6.10e7.97) h in the i.v. group (P¼0.61). There was also no significant difference in block onset time between the three groups. Conclusion: Dexamethasone 4 mg has no clinically relevant effect on the duration of sensory block provided by ropivacaine applied to the ulnar nerve.